FFMPEG return caterror

[marcorocchini]

Hi cats,

for kindness, I have this problem: I have to assign as video source for ffmpeg a .avs, and for audio source the original .mxf file that are use also by avisynth to extract the video part

I have a folder called v:\micio in wich there is a file mxf called gatto.mxf

so the audiovideo source .mxf is v:\micio\gatto.mxf (it's a xdcamhd422 file with audio similar to https://www.dropbox.com/s/7nx4i1dcql02wmv/HD169.MXF)


I try to batch automate a procedure that do a ffmpeg resize from HD to SD and FFmpeg use as source a .avs file created by the batch in the same dir of the input file (that are drag and dropped into the batch icon on the desktop, OR that are manually input typing v:\micio\gatto.mxf when request launching the batch) so the automatically .avs file created by the batch is placed in v:\micio\gatto.avs

So ffmpeg have to assume as video source v:\micio\gatto.avs

In the batch the .avs file is %2
that are gatto.avs


meanwhile the original .mxf file processed is: %CD%\%~n1.mxf
that are v:\micio\gatto.mxf

Now I have do the batch so the (red, and wrong) commandline of ffmpeg become:

v:\automazioneclip\core\ffmpeg.exe -y -i v:\micio\gatto.avs -i v:\micio\gatto.mxf -map_channel 0.1.0 -map_channel 0.2.0 -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b 256k v:\micio\gattodaHdaSDMP4.mp4

from the batch:

v:\automazioneclip\core\ffmpeg.exe -y -i "%2" -i "%CD%\%~n1.mxf" -map_channel 0.1.0 -map_channel 0.2.0 -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b 256k %WorkFolder%\%~n1daHdaSDMP4.mp4

I think the error is the -i v:\micio\gatto.mxf line, to assign the audio source, but to indicate at ffmpeg that it have to get audio from v:\micio\gatto.mxf what I have to do?

The original batch is this:

@rem unless otherwise specified for debugging reasons turn echo off
@if "%EMode%"=="" Set EMode=ON
@ECHO %EMode%
@ECHO off

SetLocal DisableDelayedExpansion

Set "EXTMXF=*.mxf"
Set "DoPop=0"
Set "ARGLVL=0"
Set "FileName-DQ="
Set VDM_LOC="v:\automazioneclip\VirtualDubMOD\virtualdubmod.exe"

set "DestFolder=."
Set "WorkFolder=%DestFolder%"

if not exist %DestFolder% mkdir %DestFolder%

rem process the arguments one by one
:ARGSMXF
	
	rem if there are no further arguments then exit the batch.
	if [%1]==[] (
		if "%ARGLVL%"=="0" (
			CALL :GetTargetNameMXF		
			EndLocal
			pause
		)
		goto :eof
	)

	rem wildcard processing
	Set "FileName=%~1"
	if "%FileName%"=="" (
		SHIFT
		goto :ARGSMXF
	)

	rem if the argument doesn't exist in the file system in any form skip it.
	if NOT EXIST %1 (
		SHIFT
		goto :ARGSMXF
	)

	if NOT [%1]==[] if NOT "%FileName:^*=%%FileName:^?=%"=="%FileName%%FileName%" (
		for %%w in (%1) DO (
			Set /a ARGLVL+=1
			CALL :ARGSMXF "%%~fw"
			Set /a ARGLVL-=1
		)
		SHIFT
		goto :ARGSMXF
	)
	Set "FileName="
		
	rem Set search/recursion variables
	Set "Pattern=%EXTMXF%"
	Set "STARTLVL=0"
	Set "FOUNDLVL=1024"
	
	rem if the argument is the current directory then process everything in it.
	if "%~f1"=="%CD%" (
		rem echo Processing Directory "%CD%"
		CALL :STARTMXF %1
		SHIFT
		goto :ARGSMXF
	)
	rem Test if the argument is a directory, if it is move to it then process everything in it, if not then move to the file's parent directory and process the file.  Then return to starting directory.
	Type NUL
	pushd "%~f1" 2> nul
	if NOT "%errorlevel%"=="0" (
		if NOT "%CD%\"=="%~dp1" (
			pushd "%~dp1"
			Set "DoPop=1"
		)
		Set Pattern="%~nx1"
	) else (
		rem echo Processing Directory "%~f1"
		Set "DoPop=1"
	)
	CALL :STARTMXF %1
	if "%DoPop%"=="1" (
		popd
		Set "DoPop=0"
	)
	SHIFT

goto :ARGSMXF

:STARTMXF <CURRENT_ARG>

	rem Count is superseded by FOUNDLVL but may still be useful for renaming purposes
	Set "Count=0"
	rem Process all files in the current directory fitting the arguments search pattern.
	for /f "delims=" %%y in ('dir %Pattern% /b /od /a-d') do (
		if /I NOT "%%~y"=="File Not Found" CALL :Process_MXF "%%~fy"
	)
	rem if it didn't find anything search all directories instead
goto :eof
	if "%Count%"=="0" if %STARTLVL% LSS %FOUNDLVL% (
		for /d %%b in (*) do (
			pushd "%%~fb"
			rem echo Processing Directory "%%~fb"
			Set /a STARTLVL+=1
			CALL :STARTMXF "%%~fb"
			Set /a STARTLVL-=1
			popd
		)
	)

goto :eof

:Process_MXF <AVI_FILENAME>

	Set /a Count+=1
	Set "FOUNDLVL=%STARTLVL%"

	echo Processing "%CD%\%~nx1"
	if /I NOT "%~x1"==".mxf" (
		echo "%~nx1"
		echo  is not a .mxf file
		rem pause
		rem goto :eof
	)
	
	)

	
	SetLocal DisableDelayedExpansion
        if exist "%~dpn1.wav" (
        set "Audio=%~dpn1.wav"
        ) else (
        set "Audio=v:\automazioneclip\system\empty.wav")
        rem Find TimeCode Info
     	Set AUCount=0;
	for /f "tokens=1,2 delims=^=" %%g in ('v:\automazioneclip\virtualdubmod\FFProbe -hide_banner -loglevel fatal -pretty -show_streams -show_entries stream^=codec_type^,width^,height^,sample_aspect_ratio:stream_disposition^=:format_tags^=timecode^,company_name^,product_name^,product_version "%~1" 2^>^&1') DO (
		 if "%%~g"=="TAG:company_name" set "T_Company_Name=%%~h"
		 if "%%~g"=="TAG:product_name" set "T_Product_Name=%%~h"
		 if "%%~g"=="TAG:product_version" set "T_Product_Version=%%~h"
		 if "%%~g"=="TAG:timecode" set "TIMECODE=%%~h"
		 if "%%~g"=="width" set "V_Width=%%~h"
		 if "%%~g"=="height" set "V_Height=%%~h"
		 if "%%~g"=="sample_aspect_ratio" set "V_PAR=%%~h"
		 if "%%~g"=="codec_type" if "%%~h"=="audio" set /a AUCount+=1
	)
	if "%V_Width%"=="" (
		echo  ERROR: Video Width NOT FOUND!
		pause
		goto :eof
	)
	if "%V_Height%"=="" (
		echo  ERROR: Video Height NOT FOUND!
		pause
		goto :eof
	)
	if "%V_PAR%"=="" (
		echo  ERROR: Video PAR NOT FOUND!
		pause
		goto :eof
	)
	rem Set "V_Q=UD"
	rem if %V_Width% LEQ 1920 if %V_Height% LEQ 1120 \\
	Set "V_Q=HD"
	rem if %V_Width% LEQ 1280 if %V_Height% LEQ 752 Set "V_Q=MD"
	if %V_Width% LEQ 1024 if %V_Height% LEQ 608 Set "V_Q=SD"
	echo  Using MXF %V_Q% Mode
	echo  PAR = %V_PAR%
	if "%TimeCode%"=="" echo Warning! no TimeCode Found!
	if NOT "%TimeCode%"=="" echo Timecode= %TIMECODE%

if "%V_Q%"=="HD" (CALL :MakeMXFHD2SD "%WorkFolder%\%~n1.vcf" "%WorkFolder%\%~n1.avs" "%WorkFolder%\%~n1.avi" "%~f1")
if "%V_Q%"=="SD" (CALL :MakeMXFSD2HD "%WorkFolder%\%~n1.vcf" "%WorkFolder%\%~n1.avs" "%WorkFolder%\%~n1.avi" "%~f1")

rem CALL :MAKEVCF "%WorkFolder%\%~n1.vcf" "%WorkFolder%\%~n1.avs" "%WorkFolder%\%~n1.avi" "%~f1"

	echo  FATTO
	echo.
	echo ---------------------------------------------
	echo.
	EndLocal
	
goto :eof


:MakeMXFHD2SD <VCF_FILENAME> <AVS_FILENAME> <AVI_FILENAME> <MXF_FILENAME>

        echo *** da File MXF HD a MP4 SD ***

     echo  *** da File MXF SD-iMX a MP4 HD e !MXF HD! ***

     echo FFVideoSource^(%4^)>%2
     echo ColorMatrix^(mode="Rec.709->Rec.601"^)>>%2
     echo IResize(720,576^)>>%2
     echo function IResize^(clip Clip, int NewWidth, int NewHeight^) {>>%2
     echo Clip>>%2
     echo SeparateFields^(^)>>%2
     echo Shift=^(GetParity() ? -0.25 : 0.25) * ^(Height()/Float(NewHeight/2)^-1.0^)>>%2
     echo E  = SelectEven^(^).Spline36resize^(NewWidth, NewHeight/2, 0,    Shift^)>>%2
     echo O  = SelectOdd^( ^).Spline36resize^(NewWidth, NewHeight/2, 0,   -Shift^)>>%2
     echo Ec = SelectEven^( ^).Spline36resize^(NewWidth, NewHeight/2, 0,  2*Shift^)>>%2
     echo Oc = SelectOdd^( ^).Spline36resize^(NewWidth, NewHeight/2, 0, -2*shift^)>>%2
     echo Interleave^(E, O^)>>%2
     echo IsYV12^(^) ? MergeChroma^(Interleave(Ec, Oc)^) : Last >>%2
     echo Weave^(^)>>%2
     echo } >>%2
     echo AssumeTFF^(^)>>%2
     echo SeparateFields^(^)>>%2
     echo Ev=SelectEven^(^).sharpen(0.20^)>>%2
     echo Od=SelectOdd^(^).sharpen^(0.20^)>>%2
     echo Interleave^(Ev,Od^)>>%2
     echo Weave^(^)>>%2
     echo AssumeFPS^(25^)>>%2

     echo processing source file %CD%\%~n1.mxf

     v:\automazioneclip\core\ffmpeg.exe -y -i "%2" -i "%CD%\%~n1.mxf" -map_channel 0.1.0 -map_channel 0.2.0 -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b 256k %WorkFolder%\%~n1daHdaSDMP4.mp4 
        

goto :eof

:MakeMXFSD2HD <VCF_FILENAME> <AVS_FILENAME> <AVI_FILENAME> <MXF_FILENAME>

     

     rem v:\automazioneclip\core\ffmpeg.exe -y -i "%CD%\%~n1.mxf" -map_channel 0.1.0 -map_channel 0.2.0 -c:v libx264 -profile:v main -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex crop=out_h=1080:y=0,scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv420p -c:a libvo_aacenc -b 256k -aspect 16:9 %WorkFolder%\%~n1EncodingDaMXForiginalePerTgRbologna420.mp4
     rem v:\automazioneclip\core\ffmpeg.exe -y -i "%CD%\%~n1.mxf" -map_channel 0.1.0 -map_channel 0.2.0 -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex crop=out_h=1080:y=0,scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -c:a libvo_aacenc -b 256k -aspect 16:9 %WorkFolder%\%~n1EncodingDaFileMXForiginale.mp4         
         

goto :eof


:GetTargetNameMXF

	Set "FileName="
	Set /p FileName=InputFileName.mxf o InputFolder - INVIO per convertire tutto (EXIT esce):
	
	rem Add or remove double-quotes if necessary

	if NOT DEFINED FileName (
		Set "FileName-DQ="
		goto :DQSkipMXF
	)
	
	Set "FileName-DQ=%FileName:"=%"
	if NOT DEFINED FileName-DQ (
		Set "FileName="
		Set "FileName-DQ="
		goto :DQSkipMXF
	)
	
	if EXIST "%FileName-DQ%" (
		if "%FileName-DQ%"=="%FileName-DQ: =%" (
			Set "FileName=%FileName-DQ%"
		) else (
			Set FileName="%FileName-DQ%"
		)
	)

:DQSkipMXF

	if /I "%FileName-DQ%"=="" Set FileName="%CD%"
	if /I "%FileName-DQ%"=="CD" Set FileName="%CD%"
	if /I "%FileName-DQ%"=="EXIT" goto :eof
	if /I "%FileName-DQ%"=="NOEXIT" goto :GetTargetNameMXF
	if DEFINED FileName (
		Set /a ARGLVL+=1
		CALL :ARGSMXF %FileName%
		Set /a ARGLVL-=1
	)
	
goto :GetTargetNameMXF

thanks

[marcorocchini]

this the ffmpeg error:

processing source file V:\micio\gatto.mxf
ffmpeg version N-64405-g7d179b1 Copyright (c) 2000-2014 the FFmpeg developers
built on Jul 3 2014 22:02:04 with gcc 4.8.3 (GCC)
configuration: --enable-gpl --enable-version3 --disable-w32threads --enable-av
isynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enab
le-iconv --enable-libass --enable-libbluray --enable-libcaca --enable-libfreetyp
e --enable-libgme --enable-libgsm --enable-libilbc --enable-libmodplug --enable-
libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libope
njpeg --enable-libopus --enable-librtmp --enable-libschroedinger --enable-libsox
r --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab -
-enable-libvo-aacenc --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx
--enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-
libxavs --enable-libxvid --enable-decklink --enable-zlib
libavutil 52. 90.101 / 52. 90.101
libavcodec 55. 68.100 / 55. 68.100
libavformat 55. 45.100 / 55. 45.100
libavdevice 55. 13.101 / 55. 13.101
libavfilter 4. 10.100 / 4. 10.100
libswscale 2. 6.100 / 2. 6.100
libswresample 0. 19.100 / 0. 19.100
libpostproc 52. 3.100 / 52. 3.100
Input #0, avisynth, from '.\gatto.avs':
Duration: 00:00:11.16, start: 0.000000, bitrate: 0 kb/s
Stream #0:0: Video: rawvideo (YUY2 / 0x32595559), yuyv422, 720x576, 25 fps,
25 tbr, 25 tbn, 25 tbc
Guessed Channel Layout for Input Stream #1.1 : mono
Guessed Channel Layout for Input Stream #1.2 : mono
Guessed Channel Layout for Input Stream #1.3 : mono
Guessed Channel Layout for Input Stream #1.4 : mono
Guessed Channel Layout for Input Stream #1.5 : mono
Guessed Channel Layout for Input Stream #1.6 : mono
Guessed Channel Layout for Input Stream #1.7 : mono
Guessed Channel Layout for Input Stream #1.8 : mono
Input #1, mxf, from 'V:\micio\gatto.mxf':
Metadata:
uid : 24271e00-2968-05c1-0270-08004602023b
generation_uid : 24271e00-2968-05c2-0270-08004602023b
company_name : SONY
product_name : Opt
product_version : 1.22
product_uid : 060e2b34-0401-0103-0e06-012002010300
modification_date: 2014-06-21 21:57:11
timecode : 01:28:04:05
Duration: 00:00:11.16, start: 0.000000, bitrate: 60493 kb/s
Stream #1:0: Video: mpeg2video (4:2:2), yuv422p(tv, bt709), 1920x1080 [SAR 1
:1 DAR 16:9], max. 50000 kb/s, 25 fps, 25 tbr, 25 tbn, 50 tbc
Stream #1:1: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:2: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:3: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:4: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:5: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:6: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:7: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
Stream #1:8: Audio: pcm_s24le, 48000 Hz, 1 channels, s32, 1152 kb/s
mapchan: invalid input file stream index #0.1

[poisondeathray]

since audio from input file1 (the mxf) , not input file0 (the avs), use

-map_channel 1.1.0 -map_channel 1.2.0

syntax for -map_channel is file.stream.channel

0.1.0 doesn't exist, because avs has no audio stream #1; Only 0.0 exists (video)

[marcorocchini]

v:\automazioneclip\core\ffmpeg.exe -y -i "%2" -i "%CD%\%~n1.mxf" -map_channel 1.1.0 -map_channel 1.2.0 -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b 256k %WorkFolder%\%~n1daHdaSDMP4.mp4

thanks poison, only 1 old problem: the output mp4 have only 1 track mono that correspond the first audio cannhel in the original .mxf
how I can modify the commandline so the output mp4 have 2 separate classic audio channel L+R stereo, in wich audio CH1 is the same of audio CH1 of the camera and audio CH2 the same of CH2 of the camera? thanks

[poisondeathray]

use filter_complex and amerge

Remember this thread?
https://forum.videohelp.com/threads/364166-audio-impossible?p=2318006

Instead of [0:1] [0:2]

You would use [1:1] [1:2] , because input file for the audio is #1 (the mxf) not #0 (the avs)

[marcorocchini]

ffmpeg.exe -y -i v:\micio\gatto.avs -i v:\micio\gatto.mxf -filter_complex "[1:1] [1:2] amerge,volume=2" -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b 256k out.mp4

I have try this but the out.mp4 have alwais the one and the first audio ch1 of the camera, I would like have 2 audio channels but I'm not able to modify the commanline

[poisondeathray]

out.mp4 have alwais the one and the first audio ch1 of the camera, I would like have 2 audio channels

Possibly your audio streams are misnumbered compared to ffmpeg's stream numbering (maybe you selected wrong , silent stream) . CH1 might not correspond to 1.1, CH2 might not correspond to 1.2 etc...

Open out.mp4 in audacity and look if there are 2 channels (one might be "silent"), 1 track . Mono would have 1 channel and 1 track

[marcorocchini]

Oh no sory, vlc confused me because I seen only "track1". Now it work well, thanks poison!

Can I answer this; and if I encode in .avi file (mpeg in avi) would be a better idea or not? thank uu

[poisondeathray]

Are you sure?




I have one of your test files "ORIGINAL.mxf". If I open up in audacity, 1st audio track is silent. The audible audio starts at track 2
0:0 is video
0:1 is 1st audio
0:2 is 2nd audio
etc...

If I do this

ffmpeg -i original.mxf -c:v copy -filter_complex "[0:1] [0:2] amerge" -c:a pcm_s16le output1.mkv

Only R audio, L is silent





But If I do this

ffmpeg -i original.mxf -c:v copy -filter_complex "[0:2] [0:3] amerge" -c:a pcm_s16le output2.mkv

It's "normal" stereo L+R (not real "stereo", but Ch2 is left and Ch3 is right, both audible)


So you see, the audio numbering might not be what you think

and if I encode in .avi file (mpeg in avi) would be a better idea or not?

MPEG2 in AVI is not good idea (not very compatible)

[marcorocchini]

I have try with -filter_complex "[1:1] [1:2]

and seems to me is all ok in out.mp4, have changel file using a .mxf that have in ch1 a microphone speek and in ch2 audiocamera, and the mp4 have correctly 1 track stereo l+r with microphone and audiocamera separate, so I think is ok

If you like I upoload on the dropbox the .mxf file

[poisondeathray]

No it's ok, as long as you're convinced the channels are mapped correctly

[marcorocchini]

Seems to me is ok: ch1 camera = ch1 mp4 and ch2 camera = ch2 mp4 but I will check better

Poison a kindness:

https://www.dropbox.com/s/u1gx85anso3acce/IMXSDAUDIO.MXF

now consider this IMX mxf (sd) that have

audio ch1 = britney from a microphjone sennh.mkh-816 (audio is better, more present)
audio ch2 = britney from audio camera (bad audio)
audio ch3 and ch4 muted (no analog source introduced, but presents)

And the commandline that convert it from .mxf to .mp4

ffmpeg.exe -y -i IMXSDAUDIO.MXF ??-filter_complex "[0:1:0] [0:1:2] amerge"?? -vf crop=720:576:0:32,scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv420p -c:v libx264 -profile:v main -level:v 3 -g 33 -bf 1 -crf 18 -flags +ilme+ildct -top 1 -x264opts tff=1 -c:a libvo_aacenc -b:a 256k -aspect 16:9 out.mp4

I cannot find the way to have correct audio on the final out.mp4: it have to be 1 track stereo L+R where is the ch1 and ch2 of the camera, not others

Many thanks

[poisondeathray]

I think it should be [0:1] [0:2]

because input file is MXF:

0:0 is video
0:1 is 1st audio
0:2 is 2nd audio
etc...

[marcorocchini]

also for me, but strangely it does't work. Consider the .mxf, for this time, is IMX SD, not HD


C:\>v:\automazioneclip\core\ffmpeg.exe -y -i v:\mxfaudio\imxSDaudio.mxf -filter_
complex "[0:1] [0:2] amerge" -vf crop=720:576:0:32,scale=interl=1:in_range=tvu
t_range=tv -pix_fmt yuv420p -c:v libx264 -profile:v main -level:v 3 -g 33 -bf 1
-crf 18 -flags +ilme+ildct -top 1 -x264opts tff=1 -c:a libvo_aacenc -b:a 256k -a
spect 16:9 v:\out.mp4
ffmpeg version N-64405-g7d179b1 Copyright (c) 2000-2014 the FFmpeg developers
built on Jul 3 2014 22:02:04 with gcc 4.8.3 (GCC)
configuration: --enable-gpl --enable-version3 --disable-w32threads --enable-av
isynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enab
le-iconv --enable-libass --enable-libbluray --enable-libcaca --enable-libfreetyp
e --enable-libgme --enable-libgsm --enable-libilbc --enable-libmodplug --enable-
libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libope
njpeg --enable-libopus --enable-librtmp --enable-libschroedinger --enable-libsox
r --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab -
-enable-libvo-aacenc --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx
--enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-
libxavs --enable-libxvid --enable-decklink --enable-zlib
libavutil 52. 90.101 / 52. 90.101
libavcodec 55. 68.100 / 55. 68.100
libavformat 55. 45.100 / 55. 45.100
libavdevice 55. 13.101 / 55. 13.101
libavfilter 4. 10.100 / 4. 10.100
libswscale 2. 6.100 / 2. 6.100
libswresample 0. 19.100 / 0. 19.100
libpostproc 52. 3.100 / 52. 3.100
Guessed Channel Layout for Input Stream #0.1 : 4.0
Input #0, mxf, from 'v:\mxfaudio\imxSDaudio.mxf':
Metadata:
uid : 7acc0e00-5468-05c1-0270-08004602023b
generation_uid : 7acc0e00-5468-05c2-0270-08004602023b
company_name : SONY
product_name : Opt
product_version : 1.41
product_uid : 060e2b34-0401-0103-0e06-012002010300
modification_date: 2014-07-16 10:46:05
timecode : 00:02:44:04
Duration: 00:00:27.40, start: 0.000000, bitrate: 62569 kb/s
Stream #0:0: Video: mpeg2video (4:2:2), yuv422p(tv), 720x608 [SAR 608:405 DA
R 16:9], max. 50000 kb/s, 25 fps, 25 tbr, 25 tbn, 50 tbc
Stream #0:1: Audio: pcm_s24le, 48000 Hz, 4 channels, s32, 4608 kb/s
Stream specifier ':2' in filtergraph description [0:1] [0:2] amerge matches no s
treams.

C:\>

[Selur]

using 'code'- or 'quote'-tags makes stuff better readable
looking at: https://trac.ffmpeg.org/wiki/AudioChannelManipulation

-af "pan=stereo:c0=c0:c1=c1"

might work (never tested)

[marcorocchini]

using 'code'- or 'quote'-tags makes stuff better readable
looking at: https://trac.ffmpeg.org/wiki/AudioChannelManipulation

-af "pan=stereo:c0=c0:c1=c1"

might work (never tested)

oh my

it works!

sorry sometimes my paw don't press well the key on keyboard ^^ thaks

[marcorocchini]

For kindness: I should have another question

Return to the HD-->SD conversion:

for example with this as source

https://www.dropbox.com/s/7nx4i1dcql02wmv/HD169.MXF

I try to encode it in SD-MP4 but in avi container, with the commandline tha assume as video source the .avs (all is well working when output container is .mp4)

ffmpeg.exe -y -i "%2" -i "%CD%\%~n1.mxf" -filter_complex "[1:1] [1:2] amerge,volume=2" -c:v libx264 -profile:v high422 -level:v 4.1 -g 33 -bf 1 -crf 18 -flags +ildct+ilme -top 1 -x264opts tff=1:colorprim=bt709:transfer=bt709:colormatrix=bt709 -filter_complex scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -aspect 16:9 -r 25 -c:a libvo_aacenc -b:a 256k workFolder%\%~n1daHDaSDMP4.avi

encoding is semi "ok" but ffmpeg report this "bitstream error startcode missing"

If I try open the .avi in virtualdub is returns the audio error



and when I open the .avi but firt frame is black (decoder is VFW "vidc.h264=x264vfw.dll, confirmed also in virtualdub File Information)

and for second: to correctly open a mpeg in avi file in virtualdub, what can I use?

I have installed the FFMPEGInputDriver 0.8.1.4 plugin but it seems act for .mp4 and not for .avi file although it's a mpeg in avi, can I configure it to open also mpeg in avi files?

[Selur]

stuffing H.264&aac inside an .avi container is a bad idea, since afaik there is no splitter which can really handle aac inside an .avi container properly (=seeking&normal playback does work).

[marcorocchini]

stuffing H.264&aac inside an .avi container is a bad idea, since afaik there is no splitter which can really handle aac inside an .avi container properly (=seeking&normal playback does work).

mm I understand, what is the best alternative? I'm tryed with mp3 lame and seems the ffmpeg don't return the "bitstream error startcode missing"



but when I open the out.avi in virtualdub I get the same error:



Is there a way to resolve?
Mediainfo says the audio part is CBR, but for VirtualDub is VBR: who have reason?

[marcorocchini]

And also: is there a plugin for virtualdub that can allow to open correctly and internally to virtualdub, mpeg-in-avi files? Possibly I would like avoiding recursion to VFW codecs, or DirecShow codecs.

I have installed in virtualdub the FFinputDriver 0.8.1.4 but it don't open mpeg-in-avi files, so virtualdub use X264VFW codec ---> = first frame black = irresolvible problem of first frame black

[marcorocchini]

using 'code'- or 'quote'-tags makes stuff better readable
looking at: https://trac.ffmpeg.org/wiki/AudioChannelManipulation

-af "pan=stereo:c0=c0:c1=c1"

might work (never tested)

Selur a kindness:

1) and to regulate volume how I can do?

2) and, to use ch3 and ch4 in place of ch1/2?

3 please my cats have an answer: can I apply your idea to a HD-mxf file? as the https://www.dropbox.com/s/7nx4i1dcql02wmv/HD169.MXF (but if you want I can upload on my dropbox anhother with audio in ch1 and ch2 because HD169.mxf don't have audio on ch1, and a minimal on ch2) hoewver I would like extract audio ch1 and ch2 of the camera and "transpose" in the final mp4. Your solution match IMX (sd) mxf file, how can I do it for HD files?

Because I can use this option: -filter_complex "[1:1] [1:2] amerge,volume=2"
but with this command volume is too high and I cannot set it to 1,5 or 1,7

Thanks

[poisondeathray]

Mediainfo says the audio part is CBR, but for VirtualDub is VBR: who have reason?

Not sure, maybe ffmpeg's lamemp3 implementation has a CBR switch, I didn't see one

Or use AC3, it's CBR only

And also: is there a plugin for virtualdub that can allow to open correctly and internally to virtualdub, mpeg-in-avi files? Possibly I would like avoiding recursion to VFW codecs, or DirecShow codecs.

I have installed in virtualdub the FFinputDriver 0.8.1.4 but it don't open mpeg-in-avi files, so virtualdub use X264VFW codec ---> = first frame black = irresolvible problem of first frame black

Did you select FFMpeg files from the dropdown menu in the open dialog box ?

Don't use AVC in AVI , or AAC in AVI both causes many problems in many programs . Especially when using b-frames. Use a different container like MP4

1) and to regulate volume how I can do?

Because I can use this option: -filter_complex "[1:1] [1:2] amerge,volume=2"
but with this command volume is too high and I cannot set it to 1,5 or 1,7

-vf volume uses decimals , not commas - so 1.5 , not 1,5

https://www.ffmpeg.org/ffmpeg-filters.html#volume

2) and, to use ch3 and ch4 in place of ch1/2?

Not sure try

-af "pan=stereo:c0=c2:c1=c3"

[marcorocchini]

-vf volume uses decimals , not commas - so 1.5 , not 1,5

Did you select FFMpeg files from the dropdown menu in the open dialog box ?

grrr

[marcorocchini]

and for the IMX case, as I have use -af "pan=stereo:c0=c0:c1=c1" how I have to change the string to regulate audio volume?


maybe:

-af "pan=stereo:c0=c0:c1=c1:volume=6db"

?

[poisondeathray]

and for the IMX case, as I have use -af "pan=stereo:c0=c0:c1=c1" how I have to change the string to regulate audio volume?


maybe:

-af "pan=stereo:c0=c0:c1=c1:volume=6db"

?

did you try

-af "pan=stereo:c0=c0:c1=c1,volume=2"

or

-filter_complex "pan=stereo:c0=c0:c1=c1,volume=2"


In ffmpeg, commas separate different filters, colons separate settings within the same filter

[marcorocchini]

I have do many catproofs:

seems to me that they are identical, and both are functional

the important things is this: by default seems to me I have to set the volume=2 to have the same levels of the original mxf

supposing this is the commanline:

ffmpeg.exe -y -i barsSD.mxf -af "pan=stereo:c0=c0:c1=c1,volume=2" -vf crop=720:576:0:32,scale=interl=1:in_range=tv:out_range=tv -pix_fmt yuv422p -c:v libx264 -profile:v high422 -level:v 3 -g 33 -bf 1 -crf 18 -flags +ilme+ildct -top 1 -x264opts tff=1 -c:a libvo_aacenc -b: 256k -aspect 16:9 out.mp4

where barsSD.mxf is the original recording of the bars and EBUL audio levels, and is this

https://www.dropbox.com/s/41k459jnbzjujt7/barsSD.MXF

I have tested the audio levels on edius, and it was a very very very difficult test because my cats get nervous when I open
erdiuscatagainst nle^^

[poisondeathray]

the important things is this: by default seems to me I have to set the volume=2 to have the same levels of the original mxf

Are you saying you must increase the audio for AAC in order to have same levels of MXF ?

[marcorocchini]

the important things is this: by default seems to me I have to set the volume=2 to have the same levels of the original mxf

Are you saying you must increase the audio for AAC in order to have same levels of MXF ?

yes bu maybe I'm in error: I'm not considered the multiple audio in edius (I am proud to not know how to use edius) mmm no no no maybe I think audio levels is "straigh" with vol=1 but I'm do others proof

[poisondeathray]

Yes, cat is probably in unfamilar environment and lost

It's close at default (no volume adjustment), but libvo_aacenc AAC encoder is not that good, and cuts of certain frequencies, so there is going to be some differences. But important point is "mean volume" is the same, and "max volume" is almost the same and isn't dangerously clipping .

You can check levels with ffmpeg, first encode the audio

ffmpeg -i barsSD.MXF -af "pan=stereo:c0=c0:c1=c1" -c:a pcm_s16le v0.wav

ffmpeg -i barsSD.MXF -af "pan=stereo:c0=c0:c1=c1" -c:a libvo_aacenc -b: 256k a0.aac
ffmpeg -i barsSD.MXF -af "pan=stereo:c0=c0:c1=c1,volume=2" -c:a libvo_aacenc -b: 256k a2.aac

Use -af volumedetect to see stats

ffmpeg -i v0.wav -vn -af "volumedetect" -f null NUL 2>v0.txt
ffmpeg -i a0.aac -vn -af "volumedetect" -f null NUL 2>a0.txt
ffmpeg -i a2.aac -vn -af "volumedetect" -f null NUL 2>a2.txt

v0
[Parsed_volumedetect_0 @ 02d27c40] n_samples: 856320
[Parsed_volumedetect_0 @ 02d27c40] mean_volume: -12.0 dB
[Parsed_volumedetect_0 @ 02d27c40] max_volume: -9.0 dB
[Parsed_volumedetect_0 @ 02d27c40] histogram_9db: 249760

a0
[Parsed_volumedetect_0 @ 02d2ad00] n_samples: 860160
[Parsed_volumedetect_0 @ 02d2ad00] mean_volume: -12.0 dB
[Parsed_volumedetect_0 @ 02d2ad00] max_volume: -8.8 dB
[Parsed_volumedetect_0 @ 02d2ad00] histogram_8db: 10946

a2
[Parsed_volumedetect_0 @ 02d2ad00] n_samples: 860160
[Parsed_volumedetect_0 @ 02d2ad00] mean_volume: -6.0 dB
[Parsed_volumedetect_0 @ 02d2ad00] max_volume: -2.8 dB
[Parsed_volumedetect_0 @ 02d2ad00] histogram_2db: 30350

[marcorocchini]

 -c:a ac3 

may help?